Improving Transportation Ops. and Mgmt through Data Integration

DataPalooza 6/3/2014 : Standards for Sharing Data Part 2

Transcript

1. Michael L. Pack, University of Maryland CATT Laboratory Improving Transportation Operations and Management Through Data Integration
2. • Many traffic management centers collecting large amounts of data: – Washington, D.C. TMC – Maryland CHART TMC – Northern Virginia TMC – Several Large County TMCs – Several Transit Agencies – Emergency Management Agencies – Computer Aided Dispatch • Coordination?!—does it exist? • How is safety and performance measured/analyzed? • How are investment decisions made? • Why isn’t there more innovation— private sector or otherwise? Transportation in and Around Washington D.C. 2
3. Making it Happen: The Early Years Agency or System #1 Agency or System #2 Agency or System #3
4. • 50 Agencies What about more Agencies?
5. 2400+ Connections
6. Architectural & Policy Issues • Agencies weren’t ready for standards • Standards weren’t ready for agencies • Cost was too high per agency • Change Management terrors • Didn’t solve MOU issues • Bandwidth issues • No archive • Etc. • Etc. • Etc.
7. Revised Mission • Still Facilitate Operations Coordination • Reduce Agency Operations Integration Costs • Reduce 3rd Party Developer Costs • Encourage and Facilitate Innovation by Others • Create an accessible transportation data archive for • Performance Reporting • Research • Operations • Planning • Decrease % of Wasted Research Dollars
8. The current method • Central-ish aggregator • 1-stop shop • Cloud-esque • NSA phone-record-style storage, processing, and analysis techniques
10. But there were still problems! • Convince agencies to share their data with other partner agencies. No problem, right?! • Convince agencies to share their data with researchers. Might have some resistance • Convince agencies to share their data with the public. Maybe more challenging…
11. How did we convince folks to share their data? • Strategy 1: Philosophical & Ethical Arguments about the greater good. • FAIL because… • Strategy 2: Legal justification. • FAIL (mostly) because… • Attempt 3: Funding. • FAIL because… • Attempt 4: Shame • Pseudo fail… • Attempt 5: Enlist an internal friend • Short term win, but long term fail…
12. The Carrot: Give them Tools & Power!!! • Provide tools to make agency data • easily accessible, • usable, and • understandable To these managers, internal end users, and ITS applications… 12 © 2013 Michael L. Pack, UMD CATT Laboratory
13. RITIS Suite of Historic Analytics Tools Incident Analysis Weather Analysis Congestion Analysis
14. Progress thus far… 14 © 2013 Michael L. Pack, UMD CATT Laboratory
15. Map View © 2013 Michael L. Pack, UMD CATT Laboratory
16. How much data are we getting today? • RITIS Today • Traffic events: 10,000 records per day: 0.001 Gb/day • Traffic detectors: 35,000,000 records per day: 5 Gb/day • Probe vehicle data: 4,200,000,000 records per day: 550 Gb/day • CCTV, weather, radio, etc: NO,STA,TSK,EPT records per day: ??? Tb/day • V2X & Automation data: ?,???,???,???,??? records per day: ??? ?b/day Our New Goal
17. Usage Statistics & Growth Plans • 3500 Website Users • Hundreds of active feed subscribers (both public & private sector) • 50% Transportation • 50% “Other” –This is VERY Important!!! • 25 State DOTs • 300+ Agencies & Private Sector Providers • Over 1-Trillion Records Downloaded in 6- months Public Feeds Rolling out this Summer Users Include: • DOTs (Federal, State, and Local) • Transit Providers • Metropolitan Planning Organizations • Emergency Management Agencies • FEMA • US Army, Air Force, Navy, Coast Guard • NorthCom • U.S. Secret Service • U.S. Capitol & Park Police • Fire & Rescue • Law Enforcement (state & local) • U.S. Joint Forces Headquarters • NSA • US Office of Personnel Management • 3rd Party Trav Info Providers • University Researchers • Consultants working on projects for the DOTs • Social Security • Pentagon Force Protection • Etc.
18. Mission Recap: Data Integration & Accessibility • Facilitate Operations Coordination • Reduce Agency Operations Integration Costs • Reduce 3rd Party Developer Costs • Encourage and Facilitate Innovation by Others • Decrease % of Wasted Research Dollars • Create an accessible transportation data archive for • Performance Reporting • Research • Operations • Planning
19. Questions & Discussion Michael L. Pack PackML@umd.edu 301.405.0722 Come see us upstairs at the Data Palooza Exhibition Area
20. Backup Slide Visualization Examples
21. Weather Forecasts © 2013 Michael L. Pack, UMD CATT Laboratory
22. Weather Alerts © 2013 Michael L. Pack, UMD CATT Laboratory
23. Radio/Scanners © 2013 Michael L. Pack, UMD CATT Laboratory
24. Sensor Data © 2013 Michael L. Pack, UMD CATT Laboratory
25. CCTV Personal Media Wall © 2013 Michael L. Pack, UMD CATT Laboratory
26. Real-time and Historic Timeline of Incident Activity © 2013 Michael L. Pack, UMD CATT Laboratory
27. Chat & File Sharing Functionality
28. Evacuation Data © 2013 Michael L. Pack, UMD CATT Laboratory
29. Staging Areas © 2013 Michael L. Pack, UMD CATT Laboratory
30. Hospitals & Beds © 2013 Michael L. Pack, UMD CATT Laboratory
31. Traffic Control Points © 2013 Michael L. Pack, UMD CATT Laboratory
32. Transit © 2013 Michael L. Pack, UMD CATT Laboratory
33. Bus Stops & AVL © 2013 Michael L. Pack, UMD CATT Laboratory
34. Virtual Weigh Stations © 2013 Michael L. Pack, UMD CATT Laboratory
35. See Details of POI © 2013 Michael L. Pack, UMD CATT Laboratory
36. RITIS Suite of Historic Analytics Tools Incident Analysis Weather Analysis Congestion Analysis
37. Safety Data Analysis
40. Incident Data Analysis
44. Congestion Analytics
45. Example 1: Statewide Reporting • You’ve been asked to provide a monthly state-wide congestion report to the Secretary. This report only needs to cover the interstates, but it needs to highlight where the worst congestion occurred (top 10 locations) and some basic stats about the severity of the congestion at each of these locations. You also need to let the Secretary know if the congestion is about the same, better, or worse than the previous 2-weeks. What do you do?
46. Bottleneck Ranking Maps
47. Example 2: I just spent $200M, and all I got was this… • You just spent $200M on a 6-month major road widening project along that corridor you (and everybody else) hate. Some commuters are now complaining that things haven’t improved--- in fact, they claim things have gotten worse. You can see the headlines now: “$200M fattens road, shrinks commuter patience!” • What can you produce to show the true impact of this recent investment (positive or negative). © 2013 Michael L. Pack, UMD CATT Laboratory
48. Answer #1: better or worse? © 2013 Michael L. Pack, UMD CATT Laboratory
49. Answer #2: show me the money (User Delay Cost) © 2013 Michael L. Pack, UMD CATT Laboratory
50. © 2013 Michael L. Pack, UMD CATT Laboratory
51. Winter Weather Worries • Snowmageddon 2011. There’s been a request from the Governor’s office to produce some examples that depict how bad traffic was during the January 26th, 2011 snow storm compared to normal weekday traffic. What can you show in just a few minutes? © 2013 Michael L. Pack, UMD CATT Laboratory
52. Trend Maps
53. Trend Maps: exporting as animated GIF (snow starts around 3PM)
54. Trend Map: comparing
55. Transit Performance
57. RITIS User Stats (as of May, 2014) Users Include: • DOTs (Federal, State, and Local) • Transit Providers • Metropolitan Planning Organizations • Emergency Management Agencies • FEMA • US Army, Air Force, Navy, Coast Guard • NorthCom • U.S. Secret Service • U.S. Capitol & Park Police • Fire & Rescue • Law Enforcement (state & local) • U.S. Joint Forces Headquarters • NSA • US Office of Personnel Management • 3rd Party Trav Info Providers • University Researchers • Consultants working on projects for the DOTs • Social Security • Pentagon Force Protection • Etc. • 3,500 Website Users • Tens of thousands of users via 3rd party applications 57